Question

I have a select statement that gathers data from one monthly-summary table like the following

companyname, employee, id, usage_a, ... usage_m 

I need to create report from 6 monthly-summary tables that looks like the following

companyname, employee, id, 
jan_usage_a, ...june_usage_a, average (jun_jan_usage_a), 
...  
jan_usage_m, ...june_usage_m, average (jan_jun_usage_m)

I have done something similar with less columns, but i had to create a table with columns for each month’s usage and then merge data in month to month with a single query. At the end I would export to excel create another column to average the each usage over the months required.

For the report above I’d have to create a table over one hundred columns.
My question is is there a better way to do this?
Thanks.

Sorry for the title if it’s a little unclear.

Answer 1

If you have six different tables, then you need to join them together:

select tjan.companyname, tjan.employee, tjan.id, <rest of tjan columns>, <rest of jfeb columns>
from tjan join
     tfeb 
     on tjan.companyname = tfeb.companyname and
        tjan.employee = tfeb.employee and
        tjan.id = tfeb.id
etc. etc. etc.

The problem that you have is that the populations in the different months may be different, so the joins will lose rows. A good way to handle this is with a driving table:

select . . .
from (select companyname, employee, id from tjan union
      select companyname, employee, id from tfeb union
      . . .
     ) driving left outer join
     tjan
     on tjan.companyname = driving.companyname and
        tjan.employee = driving.employee and
        tjan.id = driving.id left outer join
     tfeb
     on tfeb.companyname = driving.companyname and
        tfeb.employee = driving.employee and
        tfeb.id = driving.id left outer join
    . . .

You can do all this in one SQL statement. There are repetitive parts (such as the column names in the select). Consider using Excel to generate these.